English

• Metal halide perovskite nanocrystals (MHP NCs) have garnered great attention in photocatalysis due to their unique electronic energy band structure of appropriate redox potential and outstanding light absorption and charge transfer efficiency [1–4]. Recently, great catalytic performances of MHP NCs photocatalysts on water splitting, CO₂ reduction, dye and antibiotic degradation have been extensively investigated [5–8]. However, MHP NCs are sensitive to polar solvent and hyperthermia on accounts of the high delocalization activity and low lattice formation energy of surface ions, which severely hinder the photocatalytic applications [9,10]. Additionally, environmental pollution caused by antibiotic contamination is a growing problem. The chemical stability and inherent resistance of tetracycline hydrochloride (TCH) molecules have led to their widespread accumulation in aquatic environments, causing serious impacts on ecosystems and human health [11]. Therefore, it is significant to explore novel MHP NCs-based photocatalysts exhibiting high performance for TCH degradation in aqueous solution.

  Several attempts have been implemented to overcome the drawbacks of MHP NCs, involving surface modification, element doping, and heterostructure construction [12]. Among them, the synergistic effect in heterostructure holds advantages in accelerating the separation and transfer of charge carriers, and thus enables the collaboration of each component to promote light adsorption and improve degradation efficiency [13]. On this respect, MHP NCs have been widely integrated with diverse components such as Cs₄PbBr₆, CsPb₂Br₅, and Rb₄PbBr₆ at the nanoscale to create new heteronanocrystals (HNCs) with well-defined interfaces and superior properties [14–22]. Two dimensional layered CsPb₂Br₅ perovskite was considered to be a good candidate for construction of heterojunction with CsPbBr₃ NCs due to its minimal lattice mismatch [23]. Several strategies on assembly of CsPbBr₃/CsPb₂Br₅ heterostructures have been developed, in which the close contact interface in CsPbBr₃/CsPb₂Br₅ is conducive for the photocatalytic reaction on accounts of unique internal charge transfer mechanism [23–25]. However, lack of molecular enrichment of pollutants remarkably restrict the photocatalytic properties of CsPbBr₃/CsPb₂Br₅ HNCs in aqueous solution [26]. Encapsulation of CsPbBr₃/CsPb₂Br₅ HNCs in porous materials is a promising way to overcome this restriction [27,28].

  Metal-organic frameworks (MOFs) are significant crystalline materials constructed from metal ions/metal clusters with organic linkers through coordination bonding effect [29]. Zeolitic imidazolate framework (ZIF-8) is a subtype of MOFs assembled from Zn²⁺ ions and 2-methylimidazole, exhibiting outstanding chemical stability [30]. Moreover, ZIF-8 has a positive valence band, showing opportunity to assemble a Z-scheme heterojunction with MHP NCs to achieve an efficient separation of photoexcited electrons and holes [31]. However, numerous amounts of water/oxygen molecules can come through the pore structure of ZIF-8 to decompose CsPbBr₃ NCs ascribing to its incompletely isolation from the surroundings [32,33]. Coating a hydrophobic polymer on the surface of MOF to construct double-protected shells over CsPbBr₃ NCs is a promising approach to elevate the stability [34].

  In this work, the CsPbBr₃/CsPb₂Br₅ HNCs were successfully encapsulated into ZIF-8 by capitalizing on the spatial constraint effect of MOF and controlling the molar ratio of Cs⁺/Pb²⁺ precursors under a thermal injection condition. Subsequently, the surface of CsPbBr₃/CsPb₂Br₅@ZIF-8 was modified with poly(methyl methacrylate) (PMMA) polymer to form CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite. The inner CsPbBr₃/CsPb₂Br₅ HNCs with excellent optoelectronic properties can facilitate the transfer of photogenerated electrons and holes, contributing to the high photocatalytic efficiency. The ZIF-8 shell can help to enrich TCH species around the catalyst, and the outmost modification of PMMA afford the composite hydrophobicity. As a result, the CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite exhibited excellent TCH degradation efficiency (~87%) in aqueous solution within 100 min under visible light irradiation. This work provides a simple and feasible approach on assembling water-stable MHP NCs photocatalysts for effective degradation of antibiotics in aqueous solution.

  The preparation process of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite is illustrated in Fig. 1a. The encapsulation of CsPbBr₃/CsPb₂Br₅ in ZIF-8 were prepared through an in-situ hot injection method. Subsequently, CsPbBr₃/CsPb₂Br₅@ZIF-8 and PMMA powder in a certain mass ratio were fully milled and then annealed under nitrogen atmosphere at 240 ℃ (higher than the melting point of PMMA) for 30 min to obtain the CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite. Transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) were tested to characterize the internal structure of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. The TEM image in Fig. 1b shows the distribution of CsPbBr₃/CsPb₂Br₅ HNCs (red dotted circle) encapsulated in ZIF-8. From the HRTEM image shown in Fig. 1c, the ZIF-8 exhibited a clear lattice fringe with a 0.333 nm spacing corresponds to the (134) crystal plane, and the PMMA polymer layer was homogeneously coated on the surface of ZIF-8 [34]. In Fig. 1d, the lattice spacings of 0.291 and 0.238 nm for CsPbBr₃, and that of CsPb₂Br₅ (0.425, 0.300 nm) match well with the (200), (211) crystal planes of CsPbBr₃ and that of (200), (220) for CsPb₂Br₅, respectively. This further indicates the successful incorporation of CsPbBr₃/CsPb₂Br₅ HNCs into ZIF-8.

  Figure 1

  

  Figure 1.  (a) Schematic diagram for preparation of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. (b) TEM and (c, d) HRTEM of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite.

  DownLoad: Full-Size Img PowerPoint

  The morphologies of ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA were characterized by scanning electron microscopy (SEM) and the results are shown in Figs. 2a–c. The ZIF-8 exhibits an interpenetrating twin morphology with a microparticle size of < 2.0 µm (Fig. 2a). The morphology of ZIF-8 did not change after encapsulation of CsPbBr₃/CsPb₂Br₅ HNCs, demonstrating that the thermal injection process did not cause the ZIF-8 crystal structure to collapse (Fig. 2b). Then, the CsPbBr₃/CsPb₂Br₅@ZIF-8 was mixed with PMMA powder in nitrogen atmosphere and annealed for 30 min at 240 ℃ to obtain CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite. As shown in Fig. S1 (Supporting Information), the morphology of PMMA is spherical with the size of about 10 µm. After annealing treatment the spherical PMMA polymer melted and disappeared after it was uniformly coated over ZIF-8 (Fig. 2c). The corresponding schematic structure diagram of ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA are displayed in Figs. 2a–c. X-ray diffraction (XRD) patterns of ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA were recorded in Fig. 2d. The XRD peaks of ZIF-8 located at 2θ of 7.4°, 10.4°, 12.7°, 14.7°, 16.4°, 18.0° and 22.1° can be indexed into the (011), (002), (112), (022), (013), (222) and (114) crystal planes, respectively, confirming the phase purity [35]. The appearance of diffraction peaks of cubic CsPbBr₃ (JCPDS No. 54–0752) and tetragonal CsPb₂Br₅ (JCPDS No. 25–0211) in CsPbBr₃/CsPb₂Br₅@ZIF-8 composite proved the successful encapsulation of CsPbBr₃/CsPb₂Br₅ HNCs in ZIF-8. The XRD pattern of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA was consistent with that of CsPbBr₃/CsPb₂Br₅@ZIF-8, demonstrating that the PMMA polymer coating has no effect on the structure stability and crystallinity. Elemental mapping images of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA in the selected area were recorded (Figs. 2e–j), confirming the coexistence and uniform distribution of Cs, Pb, Br, Zn, N and O elements in the material.

  Figure 2

  

  Figure 2.  SEM images, corresponding schematic diagram and structure diagram of (a) ZIF-8, (b) CsPbBr₃/CsPb₂Br₅@ZIF-8 and (c) CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite. (d) XRD patterns and energy dispersive spectrometry (EDS) mapping of (e) Cs, (f) Pb, (g) Br, (h) Zn, (i) N, and (j) O, respectively.

  DownLoad: Full-Size Img PowerPoint

  Fourier transform infrared spectroscopy (FT-IR) spectra were recorded to further investigate the surface functional groups and ascertain whether PMMA was successfully coated on CsPbBr₃/CsPb₂Br₅@ZIF-8 composite. As shown in Fig. 3a, the peak located at 426 cm^-1 corresponds to Zn-N stretching vibration, and the peaks located at 995, 1147, and 1423 cm^-1 correspond to C—N stretching vibrations. The peak at 2930 cm^-1 is the aliphatic C—H stretching vibration in 2-methylimidazole and the peak at 1584 cm^-1 is related to that of C=N [36]. Additionally, the peaks at 2994 and 1737 cm^-1 can be attributed to the stretching vibrations of the C—H bond and C=O bond in PMMA, respectively. The PMMA also exhibits additional vibration bands at 1447, 751, 1245 and 1143 cm^-1 [37]. It was evident that the PMMA polymer was successfully grafted onto ZIF-8 to form CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite.

  Figure 3

  

  Figure 3.  (a) FT-IR spectra of CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. (b) N₂ adsorption-desorption isotherms of ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. XPS of (c) Pb 4f and (d) Br 3d for CsPbBr₃, CsPbBr₃/CsPb₂Br₅ and CsPbBr₃/CsPb₂Br₅@ZIF-8 samples.

  DownLoad: Full-Size Img PowerPoint

  N₂ adsorption-desorption isotherms of ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite were then measured (Fig. 3b) and the detailed data are collected in Table S1 (Supporting information). The CsPbBr₃/CsPb₂Br₅@ZIF-8 composite features a decreased BET surface area (82.514 m²/g) and pore volume (0.053 cm³/g) in comparison to that of ZIF-8 (436.315 m²/g for BET and 0.233 cm³/g for pore volume) due to the encapsulation of CsPbBr₃/CsPb₂Br₅ HNCs in ZIF-8. While the structural porosity is significantly promoted after PMMA network structure coating on CsPbBr₃/CsPb₂Br₅@ZIF-8, and the specific surface area and pore volume of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA are 162.015 m²/g and 0.097 cm³/g, respectively. Furthermore, the pore size distribution of ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8, and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite are evaluated through density functional theory (DFT), and the comparisons of data are listed in Table S1. The pore size distribution of CsPbBr₃/CsPb₂Br₅@ZIF-8 is also reduced compared to ZIF-8, attributing to the insertion of CsPbBr₃/CsPb₂Br₅ HNCs.

  X-ray photoelectron spectroscopy (XPS) was used to analyze the surface chemical states. As shown in Fig. S2 (Supporting information), for high-resolution XPS pattern of Cs 3d, no significant change was found in CsPbBr₃, CsPbBr₃/CsPb₂Br₅ and CsPbBr₃/CsPb₂Br₅@ZIF-8 samples, indicating the weak interactions between Cs⁺ and PbBr₆^4- octahedron [38]. The peak differential analysis of Pb 4f was shown in Fig. 3c. The peaks at 138.7 and 143.5 eV with high binding energies correspond to the crystal Pb²⁺ ions, while the two peaks at 136.6 and 141.4 eV with low binding energies correspond to the surface Pb²⁺ ions [39]. In the CsPbBr₃/CsPb₂Br₅ HNCs, the XPS signal of crystal Pb²⁺ ions is increased, and that of the surface Pb²⁺ ions is attenuated, demonstrating the assembly of Pb-Br octahedron. In other words, the surface Pb²⁺ ions indicated the existence of surface halide vacancies defects, while they were considerably diminished after the formation of CsPbBr₃/CsPb₂Br₅ HNCs. The Br 3d spectra also demonstrated that the encapsulation of ZIF-8 can effectively passivate the surface defects of CsPbBr₃/CsPb₂Br₅ HNCs through increasing binding energy of Pb-Br and lowering binding energy of Cs-Br in CsPbBr₃/CsPb₂Br₅@ZIF-8 (Fig. 3d) [39,18]. The XPS peak areas of Cs-Br decreases gradually from CsPbBr₃ NCs to CsPbBr₃/CsPb₂Br₅ HNCs and CsPbBr₃/CsPb₂Br₅@ZIF-8, and this phenomenon also implies the combination of Cs⁺ with discrete PbBr₆^4- to form CsPbBr₃/CsPb₂Br₅ HNCs. The Raman spectra of CsPbBr₃ and CsPbBr₃/CsPb₂Br₅ HNCs were compared in Fig. S3 (Supporting information). Two remarkable characteristic Raman vibrations of 84.8 and 144.3 cm^-1 were observed for CsPbBr₃, in which the phonon vibration mode of TO₂ at 84.8 cm^-1 is related to the vibration of [PbBr₆]^4-, whereas the broad low-intensity peak of 144.3 cm^-1 corresponds to the phonon vibration mode of TO₃, mainly associated with the motion of Cs⁺ ions [40]. Compared with CsPbBr₃, the left-shift of Raman peaks of CsPbBr₃/CsPb₂Br₅ and the enhancement of the peak at 140.1 cm^-1 indicate the transformation of CsPbBr₃ into CsPb₂Br₅, further passivating the surface defects of CsPbBr₃ [41]. These Raman signals are not significant for CsPbBr₃/CsPb₂Br₅@ZIF-8 composite due to the encapsulation of ZIF-8.

  The optical properties of CsPbBr₃ NCs, CsPbBr₃/CsPb₂Br₅ HNCs and CsPbBr₃/CsPb₂Br₅@ZIF-8 composite were investigated (Fig. 4a). The similar photoluminescence (PL) emission peak indicated that all of them had the same emission center, and compared with CsPbBr₃ NCs, CsPbBr₃/CsPb₂Br₅ HNCs has a narrower full width at half maximum and higher luminescence intensity than CsPbBr₃ NCs due to the passivation effect of CsPb₂Br₅ on halide vacancy defects. Additionally, due to the restriction of ZIF-8, the rates for photoinduced electron-hole recombination in CsPbBr₃/CsPb₂Br₅@ZIF-8 composite are restricted, supported by the corresponding weakened fluorescence intensity in comparison to CsPbBr₃/CsPb₂Br₅ HNCs [42]. Time-resolved photoluminescence (TRPL) was measured to investigate the dynamics of carrier transfer in CsPbBr₃/CsPb₂Br₅@ZIF-8 (Fig. 4b) with the detailed fitting data are listed in Table S2 (Supporting information). Average PL decay lifetimes of CsPbBr₃ NCs, CsPbBr₃/CsPb₂Br₅ HNCs, and CsPbBr₃/CsPb₂Br₅@ZIF-8 were 8.7, 36.6, and 14.1 ns, respectively. The CsPbBr₃/CsPb₂Br₅ HNCs shows an increased average PL lifetime, which implies that the formation of heterostructure passivated the non-radiative decay process of CsPbBr₃ NCs by increasing the recombination rate of carriers [24]. However, the fluorescence lifetime was reduced after encapsulation into ZIF-8, indicating that the migration rate of charge carriers was promoted in CsPbBr₃/CsPb₂Br₅@ZIF-8 composite.

  Figure 4

  

  Figure 4.  (a) PL spectra (λ_ex = 365 nm), (b) PL decays lifetimes of CsPbBr₃@ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite. (c) Water stability of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite. (d) Comparison PL intensity stable of CsPbBr₃ NCs, CsPbBr₃/CsPb₂Br₅ HNCs, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA samples in water measured at different time intervals and the corresponding optical photographs of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. (e) Schematic diagram of crystal structure and transformation process of CsPbBr₃/CsPb₂Br₅ HNCs.

  DownLoad: Full-Size Img PowerPoint

  Figs. 4c and d compared the water stability of the samples prior to the investigation of photocatalytic performance. The results showed that the luminescence of CsPbBr₃ NCs was completely quenched after soaking in water for 6 h due to the protective effect of surface hydrophobic ligands. After 16 h of immersion, the emission of CsPbBr₃/CsPb₂Br₅ HNCs was quenched because of the dissociation of CsPb₂Br₅ into CsPbBr₃ and PbBr₂ when immersed in water according to the following Eq. 1 [41]:

  (1)

  In contrast, the fluorescence intensity of CsPbBr₃/CsPb₂Br₅@ZIF-8 composite remained nearly 40% after 42 h of immersion, which was attributed to the encapsulation and protective effect of the external ZIF-8. The corresponding PL spectra of these samples after immersing in water are shown in Fig. S4 (Supporting information). Although the water stability of CsPbBr₃/CsPb₂Br₅ HNCs packaged in ZIF-8 had been significantly improved, it was not enough to cope with harsh environmental erosion in practical applications. The PMMA polymer shelled CsPbBr₃/CsPb₂Br₅@ZIF-8 not only retained the excellent hydrophobicity of the polymer and the porosity of ZIF-8 (Fig. S5 in Supporting information), but also had superior water stability with improved enrichment performance towards organic pollutants. The CsPbBr₃@ZIF-8 composite was also prepared as a reference using the same procedure (Figs. S6 and S7 in Supporting information). Lacking the protection of CsPb₂Br₅ heterostructure, the fluorescence intensity of CsPbBr₃@ZIF-8 composite decayed dramatically after modification of PMMA. In stark contrast, the PL intensity of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite was significantly enhanced compared to the initial intensity after 30 days of soaking in water due to the further passivation of CsPb₂Br₅ [34]. In detail, the PbBr₆ octahedron were formed in the process of CsPbBr₃ NCs growth, which can be transformed into PbBr₈ polyhedron in Pb-rich environment (Fig. 4e) [43,44]. Then, the PbBr₈ polyhedra shared top-angle Br atoms to produce a [Pb₂Br₅]^- layer, and further bonded with Cs⁺ to form CsPb₂Br₅. When the CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite was immersed in water, CsBr was stripped by water, and part of the CsPbBr₃ were decomposed to form CsPb₂Br₅ to further passivate the surface defects of CsPbBr₃, leading to a slight fluorescence enhancement. The characteristic XRD peaks of CsPbBr₃ and CsPb₂Br₅ remained for the composite (Fig. S8 in Supporting information), confirming its high water stability. Additionally, the intensity of the XPS signals of surface Pb²⁺ ions is weakened, while that of the crystal Pb²⁺ ions is enhanced after water treatment, indicating the reduce of surface defects in CsPbBr₃ accompanied by the formation of CsPbBr₃/CsPb₂Br₅ HNCs (Fig. S9 in Supporting information).

  The UV–vis absorption spectra in Fig. 5a showed that pure ZIF-8 can only absorb light before 440 nm with a band gap of 4.06 eV (Fig. S10 in Supporting information), wasting a lot of visible light resources. In contrast, all of CsPbBr₃@ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composites can extend the light utilization range to 520 nm. The CsPbBr₃/CsPb₂Br₅ HNCs and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA exhibit the best light absorption intensity, showing great potential for photocatalytic application. The valence band (VB) position of ZIF-8, CsPbBr₃ and CsPb₂Br₅ was estimated using the VB-XPS spectra (Fig. 5b), which can be calculated by Eq. 2:

  (2)

  where E_VB(NHE) is the VB potential of a standard hydrogen electrode, E_VB-XPS is the VB value measured by VB-XPS, and φ is the work function of the XPS analyzer (4.2 eV) [45–47]. Therefore, the E_VB(NHE) of ZIF-8, CsPbBr₃ and CsPb₂Br₅ is determined to be 3.41, 1.39 and 1.11 eV (vs. NHE), respectively. XRD patterns demonstrate the pure phase of the prepared CsPbBr₃ and CsPb₂Br₅ (Fig. S11 in Supporting information). Moreover, the photoelectrochemical test was used to investigate the charge transport efficiency of photogenerated carriers. As demonstrated in Fig. 5c, all samples displayed a consistently stable photoresponse for 10 light-on/off cycles. The photocurrent response of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA was remarkably higher than the other samples. The increased photocurrent suggests a significant reduction of the charge transfer resistance of the sample, demonstrating that the formation of CsPbBr₃/CsPb₂Br₅ HNCs and the encapsulation of ZIF-8 promoted internal carrier transport [13]. Similarly, the typical Nyquist plots from Electrochemical impedance spectroscopy (EIS) showed that the CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite has the smallest arc radius, indicating the lowest charge transfer resistance and high charge transfer rate (Fig. 5d).

  Figure 5

  

  Figure 5.  (a) UV–vis absorption spectra, (b) VB-XPS spectra of ZIF-8, CsPbBr₃ and CsPb₂Br₅. (c) Measurement of transient photocurrent in response to on-off illumination. (d) EIS Nyquist plots of catalytic agent.

  DownLoad: Full-Size Img PowerPoint

  The photocatalytic property of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA was then investigated by degradation of TCH under visible light irradiation (λ > 400 nm). Typically, 50 mL 40 mg/L TCH aqueous solution and 20–50 mg CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA were mixed in a quartz reactor. After that, the mixed solution was stirred without light for 30 min to achieve an adsorption/desorption balance between the TCH and the catalysts. An equal amount of solution was centrifuged at different time intervals to remove the photocatalyst. The pH of the photocatalytic reaction solution is close to neutral. The characteristic peaks were detected using UV–vis spectrophotometer and the concentration changes in the range of 230–500 nm were determined (Fig. S12 in Supporting information). Fig. 6a displays the concentration (C/C₀) variation plots of different photocatalysts. Before photocatalysis, the catalyst powder was magnetically stirred for 30 min to achieve the adsorption-desorption equilibrium of pollutant. Only 7.8% of TCH was self-degraded after 100 min with the absence of any photocatalyst. When ZIF-8 was added as a photocatalyst, the degradation rate reached only 33.1% in 100 min. CsPbBr₃@ZIF-8 was also tested as a reference under the same condition with a degradation rate of 44.8%. The CsPbBr₃/CsPb₂Br₅@ZIF-8 photocatalyst has a degradation efficiency of about 63.7%. Different CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composites were prepared by adjusting the mass ratio of CsPbBr₃/CsPb₂Br₅@ZIF-8 to PMMA (Fig. S13 in Supporting information). The results proved that the highest degradation rate can up to 87.2% when the mass ratio of CsPbBr₃/CsPb₂Br₅@ZIF-8 to PMMA was 1:3.

  Figure 6

  

  Figure 6.  (a) Time-dependent absorption spectra of TCH aqueous solution under visible light irradiation (λ > 400 nm, m_catalyst = 40 mg). (b) Plot of C/C₀ as a function of irradiation time using varying amounts of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA catalysts: 20, 30, 40, and 50 mg. (c) Four cycles of photo-degradation of TCH over CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. (d) The effect of several trapping reagents on photocatalytic degradation.

  DownLoad: Full-Size Img PowerPoint

  The effect of the dosage of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite on the photocatalytic degradation of TCH were further studied. With the increase of photocatalyst from 20 mg to 40 mg, the TCH degradation efficiency is also increased (Fig. 6b). Nevertheless, when the photocatalyst mass reaches up to 50 mg, the photocatalytic efficiency decreased slightly, which may be caused by the aggregation of catalyst. The cyclic stability of photocatalysts is crucial in evaluating their performance, and which was assessed by measuring the degradation of TCH over four consecutive runs. Fig. 6c demonstrates that the CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA can maintain photocatalytic activity even after 4 cycles, indicating its good durability. Moreover, we measured the XRD pattern after 5 cycles of degradation, which revealed no significant miscellaneous peaks (Fig. S14 in Supporting information). Photogenerated ^•O₂^-, holes (h⁺) and ^•OH are crucial in photodegradation. Trapping experiments were conducted to determine which radical dominates during the photocatalytic degradation of TCH to enhance the comprehension of photocatalytic mechanism (Fig. 6d). Benzoquinone (BQ), ethylenediaminetetraacetic acid disodium salt (EDTA-2Na), and tert-butanol (TBA) are selected as radical trappers for ^•O₂^-, h⁺, and ^•OH, respectively (Fig. S15 in Supporting information). The TCH degradation efficiency decreased when the quencher of EDTA and BQ were introduced, demonstrating that h⁺ and ^•O₂^- species play dominant roles in TCH degradation.

  The energy band structures of the cubic phase CsPbBr₃ and tetragonal phase CsPb₂Br₅ were calculated using DFT, and the results demonstrates that the CsPbBr₃ is a direct bandgap semiconductor with a small bandgap of 1.754 eV and the CsPb₂Br₅ is an indirect bandgap semiconductor with a large bandgap of 3.049 eV (Figs. 7a and b). The above theoretically calculated bandgap value of CsPb₂Br₅ is consistent with the experimental finding (E_g, 3.13 eV), whereas the fitted bandgap of CsPbBr₃ is slightly lower than that of the experimental E_g value (2.28 eV) (Fig. S16 in Supporting information) [24]. Accordingly, the E_CB was determined to be −0.65, −0.89 and −2.02 eV (vs. NHE) for ZIF-8, CsPbBr₃ and CsPb₂Br₅, respectively, using the equation of E_VB = E_g + E_CB. Based on the comprehensive findings from the reported literature, the findings revealed that the typical type-Ⅱ heterojunction was constructed between CsPbBr₃ and CsPb₂Br₅, and the Z-scheme heterojunction was formed between CsPb₂Br₅ and ZIF-8 (Fig. 7c) [13,42]. When high-energy light irradiated CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA, electrons in the valence bands of CsPbBr₃, CsPb₂Br₅, and ZIF-8 were excited and jumped to the conduction bands, producing photoelectrons (e^-) and leaving corresponding photoholes (h⁺) in the valence bands. Due to the construction of the type-Ⅱ heterojunction, photoelectrons on the conduction band of CsPb₂Br₅ would be transferred to CsPbBr₃. Similarly, the photoelectrons in the conduction band of ZIF-8 transferred to recombine with the holes in the valence band of CsPb₂Br₅ on accounts of the formation of Z-scheme heterojunction, which lead to separation of photogenerated electrons and holes. At the moment, the CsPbBr₃ conduction band gathered electrons and then activated O₂ to form superoxide radicals (^•O₂^-); the holes that accumulated on the valence band of ZIF-8 reacted with H₂O to form hydroxyl radicals (^•OH). The maximum separation efficiency of photoelectrons and holes was achieved in this reaction system, promoting the photocatalytic TCH degradation.

  Figure 7

  

  Figure 7.  The energy band structures of (a) CsPbBr₃ and (b) CsPb₂Br₅. (c) Mechanism diagram of the CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA photocatalyst for TCH photo-degradation.

  DownLoad: Full-Size Img PowerPoint

  In summary, we judiciously prepared an ultrastable CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite by capitalizing on the confinement effect of ZIF-8 and controlling the ratio of Cs⁺/Pb²⁺ precursors through thermal injection process, and then modifying the surface of ZIF-8 with hydrophobic polymer PMMA. Detailed characterizations confirmed the effective construction of the type-Ⅱ heterojunction between CsPbBr₃ and CsPb₂Br₅ and its accelerated charge transfer and separation. Moreover, the Z-Scheme heterostructure constructed by the introduction of ZIF-8 not only further accelerated the charge separation, but also improved the enrichment of TCH pollutants. Thanks to the unique band structure combining type-Ⅱ and Z-scheme, the obtained catalyst showed excellent visible-light-driven photodegradation performance towards TCH in aqueous solution with a high degradation efficiency of ~87%. The catalyst also exhibits exceptional stability and reusability. Furthermore, the mechanism of the photocatalytic reaction was analyzed in detail. This research provides new insights for fabricating and realizing the application of new CsPbBr₃-based photocatalyst in aqueous phase photocatalysis.

  Declaration of completing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Le Han: Writing – original draft, Methodology, Formal analysis, Data curation. Zhou Yuan: Formal analysis. Bohan Li: Data curation. Yuchi Zhang: Formal analysis. Lin Yang: Data curation. Yan Xu: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization.

  Acknowledgments

  The present work was supported by National Natural Science Foundation of China (No. 22171040), Shenyang Young and Middle-aged Science and Technology Innovation Talent Support Program (No. RC230784), Guangdong Basic and Applied Basic Research Foundation (No. 2023A1515140011), and Fundamental Research Funds for the Central Universities, China (No. N2305017).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110349.

  
  
• 1. [1]

     L. Duan, L. Hu, X. Guan, et al., Adv. Energy Mater. 11 (2021) 2100354. doi: 10.1002/aenm.202100354

  2. [2]

     L. Protesescu, S. Yakunin, M.I. Bodnarchuk, et al., Nano Lett. 15 (2015) 3692–3696. doi: 10.1021/nl5048779

  3. [3]

     K. Li, S. Li, W. Zhang, et al., J. Colloid Interface Sci. 596 (2021) 376–383. doi: 10.1016/j.jcis.2021.03.144

  4. [4]

     H. Zhou, Y. Qu, T. Zeid, et al., Energy Environ. Sci. 5 (2012) 6732–6743. doi: 10.1039/c2ee03447f

  5. [5]

     D. Cardenas-Morcoso, A.F. Gualdrón-Reyes, A.B.F. Vitoreti, et al., J. Phys. Chem. Lett. 10 (2019) 630–636. doi: 10.1021/acs.jpclett.8b03849

  6. [6]

     G. Gao, Q. Xi, H. Zhou, et al., Nanoscale 9 (2017) 12032–12038. doi: 10.1039/C7NR04421F

  7. [7]

     K. Xie, S. Wei, A. Alhadhrami, Adv. Compos. Hybrid Mater. 5 (2022) 1423–1432. doi: 10.1007/s42114-022-00520-4

  8. [8]

     J. Zheng, Y. Zhang, C. Jing, et al., Adv. Compos. Hybrid Mater. 5 (2022) 2406–2420. doi: 10.1007/s42114-022-00448-9

  9. [9]

     T. Ghosh, S. Aharon, A. Shpatz, et al., ACS Nano 12 (2018) 5719–5725. doi: 10.1021/acsnano.8b01570

  10. [10]

      J. Chen, D. Liu, M.J. Al-Marri, et al., Sci. China Mater. 59 (2016) 719–727. doi: 10.1007/s40843-016-5123-1

  11. [11]

      T. Ji, H. Zhang, S.J. Shah, et al., J. Mater. Chem. A 10 (2022) 22571–22583. doi: 10.1039/d2ta06383b

  12. [12]

      T. Liang, W. Liu, X. Liu, et al., Chem. Mater. 33 (2021) 4948–4959. doi: 10.1021/acs.chemmater.1c00542

  13. [13]

      L. Ding, B. Borjigin, Y. Li, et al., ACS Appl. Mater. Interfaces 13 (2021) 51161–51173. doi: 10.1021/acsami.1c17870

  14. [14]

      S. Bera, N. Pradhan, ACS Energy Lett. 5 (2020) 2858–2872. doi: 10.1021/acsenergylett.0c01449

  15. [15]

      A.M. Smith, S. Nie, Acc. Chem. Res. 43 (2010) 190–200. doi: 10.1021/ar9001069

  16. [16]

      G.H. Ahmed, J. Yin, O.M. Bakr, et al., ACS Energy Lett. 6 (2021) 1340–1357. doi: 10.1021/acsenergylett.1c00076

  17. [17]

      D.N. Dirin, B.M. Benin, S. Yakunin, et al., ACS Nano 13 (2019) 11642–11652. doi: 10.1021/acsnano.9b05481

  18. [18]

      G. Jiang, C. Guhrenz, A. Kirch, et al., ACS Nano 13 (2019) 10386–10396. doi: 10.1021/acsnano.9b04179

  19. [19]

      V.K. Ravi, S. Saikia, S. Yadav, et al., ACS Energy Lett. 5 (2020) 1794–1796. doi: 10.1021/acsenergylett.0c00858

  20. [20]

      J. Shamsi, Z. Dang, P. Ijaz, et al., Chem. Mater. 30 (2018) 79–83. doi: 10.1021/acs.chemmater.7b04827

  21. [21]

      G. Kaur, K. Justice Babu, et al., J. Phys. Chem. Lett. 10 (2019) 5302–5311. doi: 10.1021/acs.jpclett.9b01552

  22. [22]

      B. Wang, C. Zhang, S. Huang, et al., ACS Appl. Mater. Interfaces 10 (2018) 23303–23310. doi: 10.1021/acsami.8b04198

  23. [23]

      I. Rosa-Pardo, A. Ciccone, R. Arenal, et al., Chem. Mater. 35 (2023) 7011–7019. doi: 10.1021/acs.chemmater.3c01280

  24. [24]

      C. Zhang, Z. Wang, M. Wang, et al., ACS Appl. Mater. Interfaces 15 (2023) 35216–35226. doi: 10.1021/acsami.3c07081

  25. [25]

      P. Gao, Z. Cui, X. Liu, et al., Chem. Eur. J. 28 (2022) e202201095. doi: 10.1002/chem.202201095

  26. [26]

      P. Chen, W.J. Ong, Z. Shi, et al., Adv. Funct. Mater. 30 (2020) 1909667. doi: 10.1002/adfm.201909667

  27. [27]

      H. Huang, D. Verhaeghe, B. Weng, et al., Angew. Chem. 134 (2022) e202203261. doi: 10.1002/ange.202203261

  28. [28]

      E.M. Akinoglu, D.A. Hoogeveen, C. Cao, et al., ACS Nano 15 (2021) 7860–7878. doi: 10.1021/acsnano.0c10387

  29. [29]

      J. Li, H. Wang, X. Yuan, et al., Coord. Chem. Rev. 404 (2020) 213116. doi: 10.1016/j.ccr.2019.213116

  30. [30]

      H. Dai, X. Yuan, L. Jiang, et al., Coord. Chem. Rev. 441 (2021) 213985. doi: 10.1016/j.ccr.2021.213985

  31. [31]

      W. Mo, Z. Fan, S. Zhong, et al., Small 19 (2023) 2207705. doi: 10.1002/smll.202207705

  32. [32]

      M. He, Q. Zhang, F. Carulli, et al., ACS Energy Lett. 8 (2023) 151–158. doi: 10.1021/acsenergylett.2c02062

  33. [33]

      L. Han, Y. Han, J. Wu, et al., Mater. Chem. Front. 5 (2021) 7843–7851. doi: 10.1039/d1qm01185e

  34. [34]

      L. Han, B. Li, Y. Zhang, et al., Chem. Eng. J. 473 (2023) 145323. doi: 10.1016/j.cej.2023.145323

  35. [35]

      X.C. Yang, S.Q. Fu, Q.L. Li, et al., Chem. Eng. J. 465 (2023) 142869. doi: 10.1016/j.cej.2023.142869

  36. [36]

      A. Mohammadi, A.N. Pour, J. CO2 Util. 69 (2023) 102424.

  37. [37]

      Z.G. Ayanoğlua, M. Doğan, Diam. Relat. Mater. 108 (2020) 107950. doi: 10.1016/j.diamond.2020.107950

  38. [38]

      D. Yang, X. Li, Y. Wu, et al., Adv. Opt. Mater. 7 (2019) 1900276. doi: 10.1002/adom.201900276

  39. [39]

      W. Wang, R. Guo, X. Xiong, et al., Mater. Today Phys. 18 (2021) 100374. doi: 10.1016/j.mtphys.2021.100374

  40. [40]

      D. Han, K. Yang, C. Bai, et al., Chem. Eng. J. 475 (2023) 146209. doi: 10.1016/j.cej.2023.146209

  41. [41]

      Y. Ding, S. Maitra, C. Wang, et al., Interdiscip. Mater. 1 (2022) 213–255. doi: 10.1002/idm2.12025

  42. [42]

      Y. Hou, F. Meng, J. He, et al., J. Mater. Chem. C 11 (2023) 13570–13578. doi: 10.1039/d3tc02493h

  43. [43]

      B. Qiao, P. Song, J. Cao, et al., Nanotechnology 28 (2017) 445602. doi: 10.1088/1361-6528/aa892e

  44. [44]

      G. Li, H. Wang, Z. Zhu, et al., Chem. Commun. 52 (2016) 11296–11299. doi: 10.1039/C6CC05877A

  45. [45]

      X. Li, Q. Liu, F. Deng, et al., Appl. Cataly. B: Environ. 314 (2022) 121502. doi: 10.1016/j.apcatb.2022.121502

  46. [46]

      F. Yu, T. Huo, Q. Deng, et al., Chem. Sci. 13 (2022) 754–762. doi: 10.1039/d1sc06555f

  47. [47]

      H. Li, F. Yu, A. Li, et al., Chem. Eng. J. 489 (2024) 151219. doi: 10.1016/j.cej.2024.151219

• 

  Figure 1  (a) Schematic diagram for preparation of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. (b) TEM and (c, d) HRTEM of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  SEM images, corresponding schematic diagram and structure diagram of (a) ZIF-8, (b) CsPbBr₃/CsPb₂Br₅@ZIF-8 and (c) CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite. (d) XRD patterns and energy dispersive spectrometry (EDS) mapping of (e) Cs, (f) Pb, (g) Br, (h) Zn, (i) N, and (j) O, respectively.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) FT-IR spectra of CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. (b) N₂ adsorption-desorption isotherms of ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. XPS of (c) Pb 4f and (d) Br 3d for CsPbBr₃, CsPbBr₃/CsPb₂Br₅ and CsPbBr₃/CsPb₂Br₅@ZIF-8 samples.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) PL spectra (λ_ex = 365 nm), (b) PL decays lifetimes of CsPbBr₃@ZIF-8, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite. (c) Water stability of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA composite. (d) Comparison PL intensity stable of CsPbBr₃ NCs, CsPbBr₃/CsPb₂Br₅ HNCs, CsPbBr₃/CsPb₂Br₅@ZIF-8 and CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA samples in water measured at different time intervals and the corresponding optical photographs of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. (e) Schematic diagram of crystal structure and transformation process of CsPbBr₃/CsPb₂Br₅ HNCs.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) UV–vis absorption spectra, (b) VB-XPS spectra of ZIF-8, CsPbBr₃ and CsPb₂Br₅. (c) Measurement of transient photocurrent in response to on-off illumination. (d) EIS Nyquist plots of catalytic agent.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) Time-dependent absorption spectra of TCH aqueous solution under visible light irradiation (λ > 400 nm, m_catalyst = 40 mg). (b) Plot of C/C₀ as a function of irradiation time using varying amounts of CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA catalysts: 20, 30, 40, and 50 mg. (c) Four cycles of photo-degradation of TCH over CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA. (d) The effect of several trapping reagents on photocatalytic degradation.

  下载: 全尺寸图片 幻灯片
  

  Figure 7  The energy band structures of (a) CsPbBr₃ and (b) CsPb₂Br₅. (c) Mechanism diagram of the CsPbBr₃/CsPb₂Br₅@ZIF-8@PMMA photocatalyst for TCH photo-degradation.

  下载: 全尺寸图片 幻灯片
•